Regression for Predictive Analytics: Parametric and Nonparametric Regression

Predictive analytics plays a major role in big data. This book provides a sound statistical foundation for a broad range of topics in regression covering traditional parametric methods. The focus is on supervised learning. The authors, well respected in the field of financial engineering, follow a data-based approach complete with the necessary theory for a clear understanding of the nature and scope of the content. Derivations are provided in Technical Notes sections at the ends of chapters. Each chapter begins with two or three practical examples that pose questions of interest in the given setting. Statistical methods, which answer these questions, are, then, introduced and illustrated throughout the remainder of the chapter. Emphasis is placed on visual representations of data and, where appropriate, computer-generated plots are employed. Applications are drawn from multiple fields including business, economics, finance, marketing, engineering, computer science, and the sciences (biological, physical and social). Generic R and SAS outputs are used. An author web site, which includes further codes, instructor notes, and selected answers to over 300 exercises, is provided on a gratis basis.